Lura,
You've been hard at work. You are doing all the right things. You have
noticed that 12/12 matches with a different surname than Carroll are not
related as you compare more markers. This happened to me also - there is
one other 12/12 match to my R1a "Slavic" Carroll Y-DNA, but with a different
surname, and we concluded we were not related after comparing 25 and 37
markers. The "relationship" disappeared..
Even with a 12/12 match with the same surname there is only a 45%
probability that you are related - and of course also a 55% probability that
you are not related - within 150 years. Remember that haplogroup R1b, which
all the other Carrolls have in the Carroll Surname Project at FTDNA so far,
is very common in western Europe, and Ireland.
Someone with a good math background may have to sort all those Irish
Carrolls families out as the project matures, and matches are found.
Perhaps the different Septs can be discerned? That is for the future as the
project goes on. I was just lucky...
Best Wishes with your research.
Eric Olson
Carroll-DNA list administrator
ericbear(a)pcweb.net
----- Original Message -----
From: "Lura" <luraj(a)triad.rr.com>
To: <CARROLL-DNA-L(a)rootsweb.com>
Sent: Friday, July 01, 2005 5:44 PM
Subject: [CARROLL-DNA] Comparing Y-DNA results
I've spent the day trying to learn more about Y-DNA profiling and
what
can be learned from the results. I am still interested in trying to
understand the CARROLL families of Sampson County, NC before 1800, but
I also need the information for another surname project.
As I was doing some searching on YSearch at
http://www.ysearch.org/
this morning, I entered the ID number I had been given for one person
with the CARROLL surname who has taken the 12 marker test. I searched
for individuals who matched all 12 markers exactly and had 0 as the
genetic distance allowed. I was surprised when 60 perfect matches
came up. That means that using 12 markers as the only criteria, that
individual could "prove" a very close relationship to those 60
persons. The real shocker was that all 61 of those persons had
different surnames!
As I thought about that, I wondered how many might have been affected
by
1. Illegitimacy
2. Adoption
3. Deliberate name change
Those three just couldn't explain that many people. Then I thought
something else. In the 1600's each settler was given 50 acres of land
as a "headright" for each person he brought to Virginia. Many families
in the British Isles, especially in Catholic Ireland, wanted a better
life for their many children than was available at home. (Remember
those wives hadn't yet been granted the right to even "just say NO!")
Some were glad to send their sons (and sometimes their daughters) as
apprentices to work for wealthy land owners and learn a trade. Some of
these children assumed the name of their master by legal adoption if a
civil authority existed in the area, and some assumed the name by
convenience. Many CARROLL's with lots of children lived in Catholic
Ireland. It won't be too surprising if other surnames match.
However, more importantly, it seems to me that this search shows the
necessity of using more than 12 markers for comparison. I believe to
establish any absolute "proof" of close kinship will take the 37
marker test.
Did I do this search correctly? Have others searched using an ID with
only 12 markers and found a lot of matches?
When I searched in the same manner using the ID # for another Carroll
using the 12 markers and 0 genetic distance, I got 17 matches. None of
these perfect matches had the CARROLL surname.
With another ID # there were 6 perfect matches using the 12 makers and
0 genetic distance. Again, none of these had the Carroll surname.
However, when I used 25 or 37 markers for this individual, there were
no perfect matches.
I hope we can soon find lots of Carrolls to send a DNA sample for all
37 markers, because I believe we will need a large data base before
many matches can be made. I also hope that not many will jump to the
wrong conclusions before there are more profiles with which to
compare.
Still learning,
Lura
==== CARROLL-DNA Mailing List ====
ericbear(a)pcweb.net
==============================
Search the US Census Collection. Over 140 million records added in the
last 12 months. Largest online collection in the world. Learn more:
http://www.ancestry.com/s13965/rd.ashx